Although it is well known that some cancer patients commit suicide because of the disease, data on the suicide rate among cancer patients are very soft, only a few reports appearing, with small numbers. It is suspected that the age-adjusted rate is higher than that for the population. Comparisons with rates of patients having other illnesses are mostly anecdotal or not age corrected. It is proposed, in this study, to determine suicide rates among cancer patients from cause-of-death reports in a cohort of patients from one or more SEER cancer registries over several years. Age-adjusted cohort rates will be compared with population rates. If enough cases are available, the relationship of cancer site and suicide rate will be analyzed. Although only limited data are available for other diseases, rates for cancer patients and those with other diseases will be compared.